In cellular communication systems, the mobile station (MS) must perform timing and frequency synchronization with a base station to set up the downlink access [1]. This process is called initial cell search. As outlined by several authors [2, and references therein], the LTE initial cell search exploits two signals, namely the primary synchronization channel (P-SCH) and the secondary synchronization channel (S-SCH). The detection of these two signals enables not only time and frequency synchronization, but also provides the MS with the physical layer cell identity. In particular, physical layer cell identity can be recognized using the P-SCH, which is generated from a frequency-domain Zadoff-Chu (ZC) sequence [3]. Since the discrete Fourier transform of a ZC sequence is another ZC sequence, correct acquisition can be performed either in frequency or in time domain. One of the conventional approaches to detect the presence of a synchronization signal is represented by the power detection approach. Power detector is chosen to limit the computational costs of the decision device. In fact, the testing variable is represented by the (ordinate of the) maximum of the cross-correlation function between the received signal and its locally generated replica. The detector's output is then compared with a given threshold, often evaluated with the constant false alarm rate (CFAR) criterion. Recently, a new effective code acquisition technique based on a parabolic interpolation for spread spectrum (SS) communication systems with band-limited chip waveforms was proposed in [4]. The authors in [4] show that, via the parabolic method, it is possible to obtain a finer synchronization in the initial cell search versus the power detector approach since a parabolic fitting provides a finer testing variable by interpolating only three samples of the ambiguity function in the neighborhood of its maximum.Addressing some of these issues, this contribution proposes a time and frequency code acquisition procedure for LTE mobile networks. A related approach was proposed by the same authors to enhance the performance of time delay estimators [5]. A theoretical derivation of mean square estimation error was therein presented, carried out by a reduced Taylor's expansion up to the second order. The main contribution of our work is twofold: first, we define a new objective function that analytically expresses the performances of different detectors. In particular, we exploit the (square of the) real part and the (square of the) imaginary part of the cross-correlation function, introducing a new detector (named as super-coherent) that outperforms the conventional coherent approaches for code acquisition in 3GPP LTE networks. Linearly combining both these functions with two parametric coefficients α and β, we can achieve the performances of several detectors. The weight coefficients α and β represent percentage weights: this means that the two weights stand for how much of the real (and imaginary) part of the crosscorrelation function must be cons...