This paper investigates advanced energy-efficient wireless systems in orthogonal frequency division multiple access (OFDMA) downlink networks using coordinated multi-point (CoMP) transmissions between the base stations (BSs) in heterogeneous network (HetNet) which is adopted by 3GPP LTE-Advanced to meet IMT-Advanced targets. HetNet CoMP has received significant attention as a way of achieving spectral efficiency (SE) and energy efficiency (EE). Usually, in the literature, the total network power consumption is restricted to the sum of the power consumption of all BSs. In mobile networks, the backhaul contribution to the total power consumption is usually overlooked due to its limited impact compared to that of the radio base stations. For SE and EE analysis of HetNet CoMP, the energy and bandwidth consumption of the backhaul is considered without which the investigation remains incomplete. However, SE and EE are design criteria in conflict with each other and a careful study of their trade-off is mandatory for designing future wireless communication systems. The EE is measured as "throughput (bits) per Joule", while the power consumption model includes RF transmit, circuit and backhaul power. Furthermore, a non-ideal backhaul model such as Microwave link is also investigated within intra-HetNet-CoMP (inside one cell) where implementing fiber is not feasible. A novel inter-cell interference (ICI) coordination method is also studied to mitigate ICI. At the end, a novel resource allocation algorithm is proposed-modeled as an optimization problem-which takes into account the total power consumption, including radiated, circuit and backhaul power, and the minimum required data rate to maximize EE. Given the SE requirement, the EE optimization problem is formulated as a constrained optimization problem. The considered optimization problem is transformed into a convex optimization problem by redefining the constraint using cubic inequality, which results in an efficient iterative resource allocation algorithm. In each iteration, the transformed problem is solved by using dual decomposition with a projected gradient method. Simulations results demonstrate how backhaul has a significant impact on total power consumption and the effectiveness of the proposed schemes. In addition, the results demonstrate that the proposed iterative resource allocation algorithm converges with in a small number of iterations and illustrate the fundamental trade-offs between SE and EE. Our analytical results shed light on future "green" network planning in advanced OFDMA wireless systems such as envisioned for 5G system.