WORKING AND THEORY BEHIND THE APPLICATION

Power control:

The importance of power control MAC protocol can be explained by taking the case of fixed transmission power based and RTS/CTS based CSMA/CA protocols. Consider the four stations shown in


The transmission range of individual nodes is shown by the dashed lines. Let A wants to transmit data to B. Since C is in the listening range of sender A ( C can hear transmissions from station A) but not of receiver B, C is an “Exposed Node”. Node D is in the listening range of B (receiver) but not of A (sender). Hence, D is a “Hidden Node”. Before data transmission, A senses the channel to see if it is free. Then A sends an Request-to-Send (RTS) to B. If C hears the RTS, it defers transmission until A can hear B’s Clear-to-Send (CTS). If node B is not occupied with any other transmission, it responds back with a CTS. On hearing the CTS, node D defers its transmissions until A finishes sending data to B. When C hears a busy carrier, it defers transmission. After B receives data packet correctly, it sends back an ACK to A. The RTS/CTS handshake (as described above) creates a silence zone in which nodes have to defer their transmissions till data transmission is complete. This is a major reason for low throughput in CSMA/CA based protocols.



Suppose A wants to transmit to B and C wants to transmit to D. In the CSMA/CA MAC implementation, if C is sending to D, then A cannot send toB since B would hear the RTS from C and sense the ongoing transmission. However, if C reduced its transmission power such that it would be sufficient to reach destination node D, then it acquires a lesser floor area. So node A can also proceed with its transmission. Such a protocol closely packs the source-destination pairs in the network , which helps in other simultaneous transmission to proceed and thereby improving spatial channel reuse and hence increasing network capacity. . At any radio receiver, whether a data packet is received correctly or not is decided by the ratio of received Bit energy (Eb)-to-Total Interference per unit bandwidth I0 for the packet at the receiver which is called the Signal to-Interference (SIR) ratio. The SIR at the receiving node j is given by



where , W is the spreading bandwidth, R is the data rate , hi j is the channel gain between node i and j, Pit is the transmission power of node i and N0 is the background thermal noise. The ratio (W/R) is unity for non-spreading systems. For correct reception, received SIR should be greater than minimum threshold level SIRthres. For any ongoing transmission, the interference tolerance at a receiver j is given by rearranging Eqn as


where , ITj is the interference tolerance of node j. So, even if C is transmitting to D, A is transmitting to B and suppose D is able to listen to transmissions from A, receiver D can still decode the packet correctly if A limits its own transmission power level by considering the interference tolerance of node D. Thus, A makes its decision of transmission power level using two limiting factors: power level should be lower bounded by the value needed to reach destination B, and should be upper bounded by value which does not disrupt the ongoing transmission from C to D. This enhancement in the protocol increases the spatial channel reuse and increase spectral efficiency of the network. The basic principle in designing such a protocol would be to find an efficient mechanism that will ensure the transmission of signals with the variable power in order to satisfy the two basic requirements. First the signal must be able to reach the intended destination and second it should not interfere above the interference tolerance level at all the other nodes in the neighborhood. Based on these two information a transmitter should make a local decision on whether it can progress with the transmission or should defer since commencing its transmission can increase the interference level in the neighborhood and thus could potentially corrupt any ongoing transmission. The most difficult part in this scheme is to design the protocol in a completely distributed and self-organizing environment where there is no coordinator (like a base station in cellular network) to keep track of the interference level at each node and the distance of a receiver from a particular source.