<tr class="field"><th class="docinfo-name">Type:</th><td class="field-body">Documentary</td>
</tr>
<tr><th class="docinfo-name">Status:</th>
-<td>Draft</td></tr>
-<tr class="field"><th class="docinfo-name">TinyOS-Version:</th><td class="field-body">2.x</td>
+<td>Final</td></tr>
+<tr class="field"><th class="docinfo-name">TinyOS-Version:</th><td class="field-body">> 2.1</td>
</tr>
<tr><th class="docinfo-name">Author:</th>
<td>Rodrigo Fonseca, Omprakash Gnawali, Kyle Jamieson, and Philip Levis</td></tr>
<div class="section">
<h1><a id="abstract" name="abstract">Abstract</a></h1>
<p>The memo documents the interfaces, components, and semantics used by
-collection protocol in TinyOS 2.x. Collection provides a best-effort,
-multihop delivery of packets to the root of <em>a</em> tree. There may be
-multiple roots in a network, and in this case the semantics implemented
-are of <em>anycast</em> delivery to at least one of the roots. A node sending
-a packet does not specify which root the packet is destined to.</p>
+the collection protocols in TinyOS 2.x. Collection provides
+best-effort, multihop delivery of packets to one of a set of
+collection points. There may be multiple collection points in a
+network, and in this case the semantics are <em>anycast</em> delivery to at
+least one of the collection points. A node sending a packet does not
+specify which of the collection points the packet is destined to. The
+union of the paths from each node to one or more of the collection
+points forms a set of trees, and in this document we assume that
+collection points are the roots of these trees.</p>
</div>
<div class="section">
<h1><a id="introduction" name="introduction">1. Introduction</a></h1>
<p>Collecting data at a base station is a common requirement of sensor
-network applications. The general approach used is to build one
-or more collection <em>trees</em>, each of which is rooted at a base
-station. When a node has data which needs to be collected, it
-sends the data up the tree, and it forwards collection data that
-other nodes send to it. Sometimes, depending on the form of data
-collection, systems need to be able to inspect packets as they go
-by, either to gather statistics, compute aggregates, or suppress
-redundant transmissions.</p>
-<p>When a network has multiple base stations that act as <em>root</em> nodes,
-rather than one tree, it has a <em>forest</em> of trees. By picking a
-parent node, a collection protocol implicitly joins one of these
-trees. Collection provides a best-effort,
-multihop delivery of packets to one of a network's tree roots:
-it is an <em>anycast</em> protocol. The semantics is that the protocol
-will make a reasonable effort to deliver the message to at least
-one of the roots in the network. There are however no guarantees of
-delivery, and there can be duplicates delivered to one or more
-roots. There is also no ordering guarantees.</p>
-<p>Given the limited state that nodes can store and a general need
-for distributed tree building algorithms, simple collection protocols
-encounter several challenges. These challenges are not unique to
-collection protocols. Instead, they represent a subset of common
-networking algorithmic edge cases that occur in this protocol
-family:</p>
+network applications. The general approach used is to build one or
+more collection trees, each of which is rooted at a base station. When
+a node has data which needs to be collected, it sends the data up the
+tree, and it forwards collection data that other nodes send to
+it. Sometimes, depending on the form of data collection, systems need
+to be able to inspect packets as they go by, either to gather
+statistics, compute aggregates, or suppress redundant transmissions.</p>
+<p>Collection provides best-effort, multihop delivery of packets to one
+of a network's tree roots: it is an <em>anycast</em> protocol. The
+semantics are that the protocol will make a reasonable effort to
+deliver the message to at least one of the roots in the network. By
+picking a parent node, a node implementing the collection protocol
+inductively joins the tree its parent has joined. Delivery is best
+effort, and there can be duplicates delivered to one or more roots.
+Collection provides no ordering or real-time guarantees, although
+specific implementations may extend the basic functionality to do
+so.</p>
+<p>Given the limited state that nodes can store and a general need for
+distributed tree building algorithms, collection protocols encounter
+several challenges. These challenges are not unique to collection
+protocols. Instead, they represent a subset of common networking
+algorithmic edge cases that generally occur in wireless routing:</p>
<blockquote>
<ul class="simple">
-<li>Loop detection, detecting when a node selects one of its
-descendants as a new parent.</li>
-<li>Duplicate suppression, detecting and dealing with when lost
-acknowledgments are causing packets to replicate in the
-network, wasting bandwidth.</li>
+<li>Loop detection, for when a node selects one of its descendants as
+a next hop.</li>
+<li>Duplicate suppression, detecting and dealing with lost
+acknowledgments that can cause packets to replicate in the
+network, wasting capacity.</li>
<li>Link estimation, evaluating the link quality to single-hop
neighbors.</li>
<li>Self-interference, preventing forwarding packets along the route
from introducing interference for subsequent packets.</li>
</ul>
</blockquote>
-<p>The rest of this document describes a set of components and interfaces
-for a collection service outlined above.</p>
+<p>While collection protocols can take a wide range of approaches to
+address these challenges, the programming interface they provide is
+typically independent of these details. The rest of this document
+describes a set of components and interfaces for collection services.</p>
</div>
<div class="section">
<h1><a id="collection-interfaces" name="collection-interfaces">2. Collection interfaces</a></h1>
-<p>A node can perform four different roles in collection: producer,
-consumer, snooper, and in-network processor. Depending on their role,
-the nodes use different interfaces to interact with the collection
-component.</p>
-<p>A consumer is a root of a tree. The set of all roots and the paths that
-lead to them form the collection routing infrastructure in the network.
-For any connected set of nodes implementing the collection protocol
-there is only one collection infrastructure, <em>i.e.</em>, all roots in this
-set active at the same time are part of the same infrastructure.</p>
-<p>A node is configured to become a root by using the RootControl
-interface. RootControl.setRoot() MUST make the current node a root of
-the the collection infrastructure. RootControl.unsetRoot() MUST
-make the current root no longer a root in the collection infrastructure.
-Both calls are idempotent.
-RootControl.setRoot() MAY be called on a node that is already a root, to
-no effect. RootControl.unsetRoot() MAY be called on a node that is
-not a root:</p>
+<p>A node can perform four different roles in collection: sender (or
+source), snooper, in-network processor, and receiver (or
+root). Depending on their role, the nodes use different interfaces to
+interact with the collection component.</p>
+<p>The collection infrastructure can be multiplexed among independent
+applications, by means of a collection identifier. The collection
+identifier is used to identify different data traffic at the sender,
+intermediate-nodes, or the receiver, much like port number in TCP. All
+data traffic, regardless of the collection identifier, use the same
+routing topology.</p>
+<p>The nodes that generate data to be sent to the root are <em>senders</em>.
+Senders use the Send interface [<a class="reference" href="#id1">1</a>] to send data to the root of
+the collection tree. The collection identifier is specified as a
+parameter to Send during instantiation.</p>
+<p>The nodes that overhear messages in transit are <em>snoopers</em>. The
+snoopers use the Receive interface [<a class="reference" href="#id1">1</a>] to receive a snooped
+message. The collection identifier is specified as a parameter
+to Receive during instantiation.</p>
+<p>The nodes can process a packet that is in transit. These in-network
+<em>processors</em> use the Intercept interface to receive and update a
+packet. The collection identifier is specified as a parameter to
+Intercept during instantiation. The Intercept interface has this
+signature:</p>
+<pre class="literal-block">
+interface Intercept {
+ event bool forward(message_t* msg, void* payload, uint8_t len);
+}
+</pre>
+<p>Intercept has a single event, Intercept.forward(). A collection
+service SHOULD signal this event when it receives a packet to forward.
+If the return value of the event is FALSE, then the collection layer
+MUST NOT forward the packet. The Intercept interface allows a higher
+layer to inspect the internals of a packet and suppress it if needed.
+Intercept can be used for duplicate suppression, aggregation, and
+other higher-level services. As the handler of Intercept.forward()
+does not receive ownership of the packet, it MUST NOT modify the
+packet and MUST copy data out of the packet which it wishes to use
+after the event returns.</p>
+<p>Root nodes that receive data from the network are <em>receivers</em>. Roots
+use the Receive interface [<a class="reference" href="#id1">1</a>] to receive a message delivered by
+collection. The collection identifier is specified as a parameter to
+Receive during instantiation.</p>
+<p>The set of all roots and the paths that lead to them form the
+collection routing infrastructure in the network. For any connected
+set of nodes implementing the collection protocol there is only one
+collection infrastructure, <em>i.e.</em>, all roots in this set active at the
+same time are part of the same infrastructure.</p>
+<p>The RootControl interface configures whether a node is a
+root:</p>
<pre class="literal-block">
interface RootControl {
command error_t setRoot();
command bool isRoot();
}
</pre>
-<p>The collection infrastructure can be multiplexed among independent
-applications, by means of a <em>collection identifier</em>. It is important
-to note that the <em>data</em> traffic in the protocol is multiplexed,
-while the <em>control</em> traffic is not.</p>
-<p>The nodes that generate data to be sent to the root are <em>producers</em>.
-The producers use the Send interface [<a class="reference" href="#id1">1</a>] to send data to the root
-of the collection tree. The collection identifier is specified as a
-parameter to Send during instantiation.</p>
-<p>Root nodes that receive data from the network are <em>consumers</em>. The
-consumers use the Receive interface [<a class="reference" href="#id1">1</a>] to receive a message
-delivered by collection. The collection identifier is specified
-as a parameter to Receive during instantiation.</p>
-<p>The nodes that overhear messages in transit are <em>snoopers</em>. The
-snoopers use the Receive interface [<a class="reference" href="#id1">1</a>] to receive a snooped
-message. The collection identifier is specified as a parameter
-to Receive during instantiation.</p>
-<p>The nodes can process a packet that are in transit. These in-network
-<em>processors</em> use the Intercept interface [<a class="reference" href="#id1">1</a>] to receive and update
-a packet. The collection identifier is specified as a parameter
-to Intercept during instantiation.</p>
+<p>The first two commands MUST return SUCCESS if the node is now in the
+specified state, and FAIL otherwise. For example, if a node is already
+a root and an application calls RootControl.setRoot(), the call will
+return SUCCESS. If setRoot() returns SUCCESS, then a subsequent call
+to isRoot() MUST return TRUE. If unsetRoot() returns SUCCESS, then a
+subsequent call to isRoot() MUST return FALSE.</p>
</div>
<div class="section">
<h1><a id="collection-services" name="collection-services">3 Collection Services</a></h1>
}
}
</pre>
-<p>CollectionC MAY have additional interfaces, but they MUST have
-default functions on all outgoing invocations (commands for uses,
-events for provides) of those interfaces so that it can operate
-properly if they are not wired.</p>
+<p>CollectionC MAY have additional interfaces. All outgoing invocations
+(commands for uses, events for provides) of those interfaces MUST have
+default functions. Those default functions enable CollectionC to
+operate properly even when the additional interfaces are not wired.</p>
<p>Components SHOULD NOT wire to CollectionC.Send. The generic
component CollectionSenderC (described in section 3.1) provides
a virtualized sending interface.</p>
protocol operating on top of collection, in the same way that
different am_id_t values represent different protocols operating on
top of active messages. All packets sent with a particular
-collection_id_t generally have the same payload format, so that
-snoopers, intercepters, and receivers can parse it properly.</p>
-<p>Receive.receive MUST NOT be signaled on non-root
-nodes. CollectionC MAY signal Receive.receive on a root node when
-a data packet successfully arrives at that node. If a root node calls
-Send, CollectionC MUST treat it as it if were a received packet.
-Note that the buffer swapping semantics of Receive.receive, when
-combined with the pass semantics of Send, require that CollectionC
-make a copy of the buffer if it signals Receive.receive.</p>
-<p>If CollectionC receives a data packet to forward and it is not a
-root node, it MAY signal Intercept.forward.</p>
-<p>If CollectionC receives a data packet that a different node
-is supposed to forward, it MAY signal Snoop.receive.</p>
+collection_id_t generally SHOULD have the same payload format, so that
+snoopers, intercepters, and receivers can parse them properly.</p>
+<p>ColletionC MUST NOT signal Receive.receive on non-root
+nodes. CollectionC MUST signal Receive.receive on a root node when a
+unique (non-duplicate) data packet successfully arrives at that
+node. It MAY signal Receive.receive when a duplicate data packet
+successfully arrives. If a root node calls Send, CollectionC MUST
+treat it as it if were a received packet. Note that the buffer
+swapping semantics of Receive.receive, when combined with the pass
+semantics of Send, require that CollectionC make a copy of the buffer
+if it signals Receive.receive.</p>
+<p>If CollectionC receives a data packet to forward and it is not a root
+node, it MAY signal Intercept.forward. CollectionC MAY signal
+Snoop.receive when it hears a packet which a different node is
+supposed to forward. For any given packet it receives, CollectionC
+MUST NOT signal more than one of the Snoop.receive, Receive.receive,
+and Intercept.forward events.</p>
<p>RootControl allows a node to be made a collection tree root.
CollectionC SHOULD NOT configure a node as a root by default.</p>
<p>Packet and CollectionPacket allow components to access collection
</div>
</div>
<div class="section">
-<h1><a id="implementation" name="implementation">4 Implementation</a></h1>
-<p>An implementation of this TEP can be found in
-<tt class="docutils literal"><span class="pre">tinyos-2.x/tos/lib/net/ctp</span></tt> and <tt class="docutils literal"><span class="pre">tinyos-2.x/tos/lib/net/le</span></tt>, in
-the CTP protocol. It is beyond the scope of this document to fully
-describe CTP, but we outline its main components. CTP will be
-described in an upcoming TEP [<a class="reference" href="#id2">2</a>]. This implementation is a
-reference implementation, and is not the only possibility. It
-consists of three major components, which are wired together to form
-a CollectionC: LinkEstimatorP, CtpTreeRoutingEngineP, and
-CtpForwardingEngineP.</p>
-<p>This decomposition tries to encourage evolution of components and
-ease of use through modularization. Neighbor management and link
-estimation are decoupled from the routing protocol. Furthermore, the
-routing protocol and route selection are decoupled from the
-forwarding policies, such as queueing and timing.</p>
-<div class="section">
-<h2><a id="linkestimatorp" name="linkestimatorp">4.1 LinkEstimatorP</a></h2>
-<p>LinkEstimatorP estimates the quality of link to or from each
-neighbor. Link estimation can be done in a variety of ways, and we
-do not impose one here. It is decoupled from the establishment of
-routes. There is a narrow interface -- LinkEstimator -- between the
-link estimator and the routing engine. The one requirement is that
-the quality returned is standardized. A smaller return value from
-LinkEstimator.getQuality(), LinkEstimator.getforwardQuality(),
-LinkEstimator.getReverseQuality() MUST imply that the link to the
-neighbor is estimated to be of a higher quality than the one that
-results in a larger return value. The range of value SHOULD be
-[0,255] and the variation in link quality in that range SHOULD be
-linear. Radio provided values such as LQI or RSI, beacon based link
-estimation to compute ETX, or their combination are some possible
-approaches to estimating link qualities.</p>
-<p>LinkEstimatorP MAY have its own control messages to compute
-bi-directional link qualities. LinkEstimatorP provides calls
-(txAck(), txNoAck(), and clearDLQ()) to update the link estimates
-based on successful or unsuccessful data transmission to the
-neighbors.</p>
-<p>The user of LinkEstimatorP can call insertNeighbor() to manually
-insert a node in the neighbor table, pinNeighbor() to prevent a
-neighbor from being evicted, and unpinNeighbor() to restore eviction
-policy:</p>
-<pre class="literal-block">
-typedef uint16_t neighbor_table_entry_t
-
-LinkEstimatorP {
- provides {
- interface StdControl;
- interface AMSend as Send;
- interface Receive;
- interface LinkEstimator;
- interface Init;
- interface Packet;
- interface LinkSrcPacket;
- }
-}
-
-interface LinkEstimator {
- command uint8_t getLinkQuality(uint16_t neighbor);
- command uint8_t getReverseQuality(uint16_t neighbor);
- command uint8_t getForwardQuality(uint16_t neighbor);
- command error_t insertNeighbor(am_addr_t neighbor);
- command error_t pinNeighbor(am_addr_t neighbor);
- command error_t unpinNeighbor(am_addr_t neighbor);
- command error_t txAck(am_addr_t neighbor);
- command error_t txNoAck(am_addr_t neighbor);
- command error_t clearDLQ(am_addr_t neighbor);
- event void evicted(am_addr_t neighbor);
-}
-</pre>
-</div>
-<div class="section">
-<h2><a id="ctproutingenginep" name="ctproutingenginep">4.2 CtpRoutingEngineP</a></h2>
-<p>CtpRoutingEngineP is responsible for computing routes to the roots of a
-tree. In traditional networking terminology, this is part of the
-control plane of the network, and is does not directly forward any
-data packets, which is the responsibility of CtpForwardingEngine.
-The main interface between the two is UnicastNameFreeRouting.</p>
-<p>CtpRoutingEngineP uses the LinkEstimator interface to learn
-about the nodes in the neighbor table maintained by LinkEstimatorP and
-the quality of links to and from the neighbors. The routing protocol
-on which collection is implemented MUST be a tree-based routing
-protocol with a single or multiple roots. CtpRoutingEngineP
-allows a node to be configured as a root or a non-root node
-dynamically. CtpRoutingEngineP maintains multiple candidate next hops:</p>
-<pre class="literal-block">
-generic module CtpRoutingEngineP(uint8_t routingTableSize,
- uint16_t minInterval,
- uint16_t maxInterval) {
- provides {
- interface UnicastNameFreeRouting as Routing;
- interface RootControl;
- interface CtpInfo;
- interface StdControl;
- interface CtpRoutingPacket;
- interface Init;
- }
- uses {
- interface AMSend as BeaconSend;
- interface Receive as BeaconReceive;
- interface LinkEstimator;
- interface AMPacket;
- interface LinkSrcPacket;
- interface SplitControl as RadioControl;
- interface Timer<TMilli> as BeaconTimer;
- interface Timer<TMilli> as RouteTimer;
- interface Random;
- interface CollectionDebug;
- interface CtpCongestion;
- }
-}
-</pre>
-<pre class="literal-block">
-interface UnicastNameFreeRouting {
- command am_addr_t nextHop();
-
- command bool hasRoute();
- event void routeFound();
- event void noRoute();
-}
-</pre>
-</div>
-<div class="section">
-<h2><a id="ctpforwardingenginep" name="ctpforwardingenginep">4.3 CtpForwardingEngineP</a></h2>
-<p>The CtpForwardingEngineP component provides all the top level interfaces
-(except RootControl) which CollectionC provides and an application
-uses. It deals with retransmissions, duplicate suppression, packet
-timing, loop detection, and also informs the LinkEstimator of the
-outcome of attempted transmissions.:</p>
-<pre class="literal-block">
-generic module CtpForwardingEngineP() {
- provides {
- interface Init;
- interface StdControl;
- interface Send[uint8_t client];
- interface Receive[collection_id_t id];
- interface Receive as Snoop[collection_id_t id];
- interface Intercept[collection_id_t id];
- interface Packet;
- interface CollectionPacket;
- interface CtpPacket;
- interface CtpCongestion;
- }
- uses {
- interface SplitControl as RadioControl;
- interface AMSend as SubSend;
- interface Receive as SubReceive;
- interface Receive as SubSnoop;
- interface Packet as SubPacket;
- interface UnicastNameFreeRouting;
- interface Queue<fe_queue_entry_t*> as SendQueue;
- interface Pool<fe_queue_entry_t> as QEntryPool;
- interface Pool<message_t> as MessagePool;
- interface Timer<TMilli> as RetxmitTimer;
- interface LinkEstimator;
- interface Timer<TMilli> as CongestionTimer;
- interface Cache<message_t*> as SentCache;
- interface CtpInfo;
- interface PacketAcknowledgements;
- interface Random;
- interface RootControl;
- interface CollectionId[uint8_t client];
- interface AMPacket;
- interface CollectionDebug;
- }
-}
-</pre>
-<p>CtpForwardingEngineP uses a large number of interfaces, which can be
-broken up into a few groups of functionality:</p>
-<blockquote>
-<ul class="simple">
-<li>Single hop communication: SubSend, SubReceive, SubSnoop,
-SubPacket, PacketAcknowledgments, AMPacket</li>
-<li>Routing: UnicastNameFreeRouting, RootControl, CtpInfo,
-CollectionId, SentCache</li>
-<li>Queue and buffer management: SendQueue, MessagePool,
-QEntryPool</li>
-<li>Packet timing: Random, RetxmitTimer</li>
-</ul>
-</blockquote>
-</div>
+<h1><a id="implementation" name="implementation">4. Implementation</a></h1>
+<p>Implementations of collection can be found in
+<tt class="docutils literal"><span class="pre">tinyos-2.x/tos/lib/net/ctp</span></tt> and <tt class="docutils literal"><span class="pre">tinyos-2.x/tos/lib/net/lqi</span></tt>.
+The former is the Collection Tree Protocol (CTP), described in TEP 123
+[<a class="reference" href="#id2">2</a>]. The latter is a TinyOS 2.x port of MultihopLqi, a
+CC2420-specific collection protocol in TinyOS 1.x.</p>
</div>
<div class="section">
<h1><a id="author-addresses" name="author-addresses">5. Author Addresses</a></h1>
<table class="docutils footnote" frame="void" id="id1" rules="none">
<colgroup><col class="label" /><col /></colgroup>
<tbody valign="top">
-<tr><td class="label"><a name="id1">[1]</a></td><td>TEP 116: Packet Protocols</td></tr>
+<tr><td class="label"><a name="id1">[1]</a></td><td>TEP 116: Packet Protocols.</td></tr>
</tbody>
</table>
<table class="docutils footnote" frame="void" id="id2" rules="none">
<colgroup><col class="label" /><col /></colgroup>
<tbody valign="top">
-<tr><td class="label"><a name="id2">[2]</a></td><td>TEP 123: The Collection Tree Protocol (CTP)</td></tr>
+<tr><td class="label"><a name="id2">[2]</a></td><td>TEP 123: The Collection Tree Protocol (CTP).</td></tr>
</tbody>
</table>
</div>
projects / tinyos-2.x.git / blobdiff