1 | | ydd |
| 1 | = ALOHA with Collision Resolution (ALOHA-CR): Theory and Software Defined Radio Implementation = |
| 2 | |
| 3 | Authors: X. Liu, J.Kountouriotis, A.P. Petropulu, K.R. Dandekar |
| 4 | |
| 5 | == Abstract == |
| 6 | A cross-layer scheme, namely ALOHA With Collision Resolution (ALOHA-CR), is proposed for |
| 7 | high throughput wireless communications in a cellular scenario. Transmission occurs in a time-slotted |
| 8 | ALOHA-type fashion with a caveat: simultaneous transmissions of two users can be successful. The |
| 9 | wireless channel is assumed to be flat fading and constant over the duration of one time slot. A user i |
| 10 | with a non-empty queue transmits a packet with some probability p in the beginning of each time slot, |
| 11 | after waiting for a random time interval |
| 12 | i. If more than two users transmit in the same slot, the collision |
| 13 | cannot be resolved, the packets are discarded and the users are asked to retransmit at a later time. If only |
| 14 | one user transmits, the transmitted packet is recovered with some probability, depending on the state |
| 15 | of the channel. If two users transmit, the collision can be separated and the packets recovered by first |
| 16 | oversampling the collision signal and then exploiting independent information about the two users that |
| 17 | is contained in the signal polyphase components. The properties of the user delays |
| 18 | i’s are determined |
| 19 | so that the probability of user separation is maximized. The ALOHA-CR throughput is derived under |
| 20 | the infinite backlog assumption, i.e., the network users always have data in their queues, and also under |
| 21 | the assumption of finite backlog. In the infinite backlog assumption, the optimal contention probability |
| 22 | is calculated, which maximizes the system throughput. In the finite backlog assumption, the region of |
| 23 | contention probability is determined based on the packet arrival rate. Queuing delay of network users |
| 24 | are also derived. The performance of ALOHA-CR is demonstrated on a WARP testbed containing five |
| 25 | software defined radio nodes. The testbed results indicate ALOHA-CR leads to significant increase in |
| 26 | throughput and reduction of service delays. |
| 27 | {{{ |
| 28 | {Liu:2010, |
| 29 | Author = {X. Liu, J. Kountouriotis, A.P. Petropulu, K.R. Dandekar}, |
| 30 | Title = {ALOHA With Collision Resolution (ALOHA-CR): Theory and Software Defined Radio Implementation}, |
| 31 | Year = {2010}} |