Considering the code division multiple access (CDMA) radio communication schemes, this paper proposes a wireless access infrastructure and a protocol for factory automation with several mobile terminals. A factory manufacturing floor usually forms several islands of automation with each having multiple mobile terminals. Each island of automation can be covered by a base station that then connects to a supervisory computer (command center). The base station can then serve as a bridge between the wireless network and the MAP backbone. Adding wireless access capability to a MAP-based network can release the problem of rewiring the moving devices in floor planning and restructuring. Considering a priority structure, we propose a Handshake-Aloha protocol suitable for such wireless factory environment. The numerical results show that it performs much better than the random access schemes such as Aloha protocol in terms of the throughput and delay performance.

This paper proposes a performance evaluation approach for integrated voice and data services on CDMA wireless LANs. For random access schemes, dynamic behavior of a CDMA channel implies that we must be cautious in choosing an evaluation model which has to include the effect of dynamic error rates. The approach shown in this paper is to employ a Poisson error model which takes the occurrence of errors as a Poisson process and the occurring rate is dependent on the channel traffic. Based on this assumption, a continuous-time Markov chain model can be further developed to evaluate the throughput and delay performance. This approach can certainly apply to evaluation of many random access protocols.

A receiver-based handshake (RBHS) protocol for multipacket channels, such as code-division multiple-access (CDMA) networks, is proposed and its throughput and delay performance is analyzed. This scheme is based on a sensing period and a handshake procedure which can avoid collision and eliminates unwanted transmissions. Several system models under different assumptions, such as completely orthogonal codes and very large processing gain (we call it perfect channel) or with quasiorthogonal codes (nonperfect channel), are investigated. We assume the channel error probability due to other users' interference to be a function of the number of transmissions in a given slot as well as other system parameters. Under a slotted structure with uniform traffic assumption, the system is modeled as a discrete-time two-dimensional Markov chain. The analysis, based on various channel conditions, demonstrates that the average throughput and delay are significantly improved over other spreading code protocols, especially when the channel is likely to accommodate more channel errors, and when the handshaking time is reasonably short compared to the average packet length.>