An adaptive decode-and-forward (DF) cooperative diversity system based on quadrature modulation is proposed, in which each user is allowed to transmit its own and the partner’s data simultaneously via the in-phase and quadrature components of M -ary quadrature amplitude modulation (MQAM) constellation. Cooperative transmission in this way can avoid the bandwidth expansion in traditional cooperative systems. The closed form expression of the average bit error rate (BER) performance is derived in terms of the interuser and uplink channel statistics and a scalable power allocation factor, and further verified by simulations. Then the effect of equal power allocation (EPA) and optimal power allocation (OPA) algorithms on the system performance and the best partner location for a given user are investigated. Simulations show that the OPA algorithm offers considerable gains over EPA algorithm under practical non-ideal cooperative scenarios, and by joint consideration of partner selection and OPA, a simple yet near-optimal partner selection heuristic is achieved. An adaptive decode-and-forward (DF) cooperative diversity system based on quadrature amplitude modulation is proposed, where each user is allowed to transmit its own and the partner’s data simultaneously via the in-phase and quadrature components of M-quadrature amplitude modulation (MQAM) constellation. Cooperative transmission in this way can avoid the bandwidth expansion in traditional cooperative systems. The closed form expression of the average bit error rate (BER) performance is derived in terms of the interuser and uplink channel statistics and a scalable power allocation factor, and further verified by simulations. Then the effect of equal power allocation (EPA) and optimal power allocation (OPA) algorithms on the system performance and the best partner location for a given user are investigated. Simulations show that the OPA gains over EPA algorithm under practical non-ideal cooperative scenarios, and by joint consideration of partner selection a nd OPA, a simple yet near-optimal partner selection heuristic is achieved.